International Journal of Trend in Scientific
Research and Development (IJTSRD)
International Open Access Journal
ISSN No: 2456 - 6470 | www.ijtsrd.com | Volume - 2 | Issue – 1
A Novel Technique oof Power Extraction from
rom
A Conventional Fluorescent Lamp
Parantap Nandi
Calcutta University, Kolkata, West Bengal
ABSTRACT
Fluorescent lamps are commonly used in our homes
as a source of light. The conventional lamps use a
choke and a starter arrangement for starting discharge
in the tube. Under normal working conditions an
appreciable potential drop occurs across the choke.
The
he circuit being inductive the power factor is also
low. Now is there any way of utilizing this voltage
drop? Experiments were carried out with a view to
using this potential drop to provide energy at low
wattage without hampering the normal operation of
the
he tube light. Interesting results were obtained. The
power factor was also found to improve. In this paper
a method of utilizing the voltage drop across the
choke of a fluorescent lamp as a source of energy has
been elaborated.
Keywords: Choke; Power factor; Energy; Tube Light
Introduction
In the present day civilization, the energy crisis is ever
increasing in leaps and bounds. The need of using
energy efficient devices is undeniable. But one hardly
ponders over the process of saving/utilizing energy
from the existing equipments themselves. A very
good instance is a fluorescent lamp commonly
referred to as ‘tube light’. In addition to the lamp
itself and its holder, it requires a choke or ballast and
a make-break
break switch commonly referred to as starte
starter.
Before proceeding further a brief description of the
starting and working of a tube light is essential. When
the switch is made ON the full voltage is applied
through the ballast and starter which establishes glow
discharge in the starter. As a result off the discharge in
the starter a current is established this causes a
potential drop across the choke which tends to
extinguish the discharge. As the current gets
interrupted, the choke develops an inductive kick
given by the relation e=Ldi/dt. This high voltage
comes across the tube electrodes and strikes the
penning mixture. Gas discharge inside the tube is
initiated inside the tube and current starts flowing.
The starter becomes inactive.
These days’ fluorescent lamp installations come with
electronic ballasts which eliminate the need for a
starter and are what we call ‘instant start’. These
ballasts are undoubtedly energy saving and have a
high power factor (≥.9).
But the existing fixtures still use magnetic ballasts and
starters (the operating principle given above). The
power factor of these ballasts is quite low compared
with electronic ballasts. The tubes fitted with
magnetic ballasts and starters show two tendencies:—
tendencies:
I.
With aging, the lumen output of the lamp
decreases and the lamp gets dimmer.
dimm Finally the
lamp refuses to start and only flickers as the
production of a steady discharge becomes
impossible. The only choice is to replace the old
tube.
II. When the supply voltage is low, the magnetic
ballast is unable to produce steady discharge in
thee lamp. It continues to flicker. In this case
mostly the magnetic ballast is replaced by an
electronic one.
The above solutions rely on the ‘replacement
‘
of an
existing device in good working condition’.
condition
So
basically we abandon existing technology and resort
to new ones.
@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 1 | Nov-Dec
Dec 2017
Page: 50
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
Now is there any way by which we need not replace
an existing device but still get maximum output? With
a view to finding the answer, experiments were
conducted on a conventional fluorescent lamp using
magnetic ballast and interesting results were obtained.
Materials and apparatus required:
1. Fluorescent lamp (tube light) 36/40W.
2. Copper choke.
3. Starter
4. Transformer 230/9V, 450mA.
5. Sensitive digital multimeter
6. Incandescent lamps (standard wattages)
7. 4W LED lamp
8. 14W CFL lamp
9. Power factor meter
10. 5V battery charger
11. D.C motor (fan)
12. Capacitor 2.5µF (used with ceiling fans)
13. 1N5408 diodes
Load in Watts
0
4
5.3
14
15
40
60
100
Experimental: A conventional tube light working
using a magnetic ballast and starter was used in this
experiment. During the next sections the magnetic
ballast has been referred to as choke. When a tube
light is in the normal working condition, the choke
acts as a current limiting agent and the starter remains
inactive. It is evident from the fact that on opening the
starter the normal working of the light is unaffected.
So there are two devices from which power might be
extracted i.e. the open starter terminals and the choke
terminals. Keeping these in mind the several
experiments were conducted which have been
elaborated in the succeeding parts.
 Experiment no. 1
A load was connected across the choke of a tube light.
The remaining circuit was left as it is. The load was
varied in steps from zero to 100W. The following
table was tabulated:—
Table no. 1
Voltage across choke-load
Voltage across open starter terminal
combination in Volts
in Volts
170
125
155
123
150
115
145
118
155
120
140
110
137.5
110
130
105
A curve plotted gives a better understanding of the phenomena:—
Graph no.1
@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 1 | Nov-Dec 2017
Page: 51
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
The trend line (shown in black) explains the operating characteristics quite well. A load beyond 100W was not
used because it would increase the filament voltage beyond the tolerable limit and might damage the tube.
A linear trend line indicates the presence of a ‘maximum load’ at which voltage across the choke should
become zero. But this would mean that the parallel combination of load and choke acts as short circuit. If such
were the case two things would occur:
I.
The current limiting action of the choke would be lost and the fluorescent lamp would be subjected to
full supply voltage which would immediately destroy it.
II.
The tube light may not strike at all.
Careful experiments showed that a maximum of 150W could be applied across the choke and this too depends
on the condition of the tube. For an old tube light a maximum of 125W may be applied.
Graph no.2
Voltage across open starter terminals
Vs load curve
130
125
120
Potential drop
115
in Volts
110
105
y = -0.1291x + 117.44
100
0
20
40
60
80
100
120
Load in Watts
As the load is increased the tube light glows brighter which is visible to the naked eye. This was also measured
by appropriate arrangements. The following curve illustrates the same.
Graph no.3
It is evident that the light output increases linearly with increase in load across the choke.
@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 1 | Nov-Dec 2017
Page: 52
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
 Experiment no.2
The voltage across the choke was rectified using a bridge rectifier (making use of 4 diodes). A plot of D.C
terminal voltage Vs load current was plotted.
Graph no. 4
Again it is evident that the terminal voltage decreases
linearly with load current. An interesting thing was
observed on connecting a D.C motor at the rectifier
terminal. The tube light started flickering at the ends
and the frequency of the flickers was equal to the
frequency of rotation of the shaft.
As the D.C output current increased the tube light
grew brighter.
 Experiment no.3
The H.V side of a 230-9V, 450mA transformer was
connected in parallel with the choke. The transformer
has considerable inductance. When connected in
parallel with the choke Leffective=
. So the overall
inductance decreases. In the steady state the voltage
across the transformer primary is 137.5V while that
across the secondary is 6.4V. On shorting the L.V
side, there is no change in the primary voltage and
consequently no change in the filament voltage of the
tube light. So the filament voltage is maintained
independent of the load on the transformer. This
proves that power can be extracted at low voltage,
variable current using a transformer in parallel with
the choke.
 Experiment no. 4
The best possible way to extract power is perhaps
charging of a battery. The voltage across the choke is
capable of charging a 5V battery using an auxiliary
charger whose input is the choke voltage. This is yet
another way of obtaining energy at a low voltage from
the choke.
 Experiment no.5
With a view to improving power factor, a capacitor of
2.5µF used in ceiling fans is connected across the
choke. But it has a detrimental effect; it does not
allow the tube light to strike. This may be explained
in the following way:
At the time of starting the function of the choke is to
produce a high voltage impulse given by the relation
e=Ldi/dt which is necessary for ionization the gas
inside the tube at low pressure. Here the basic
property of an inductor which is to oppose the rate of
change of current is utilized. This is done by the
production of a voltage impulse. But a capacitor
opposes change in voltage. Since it is connected in
parallel with the inductor, it tends to oppose the high
voltage impulse produced by the inductor. This in turn
results in a current given by i=Cdv/dt. When the
switch is put on the current in the inductor tends to
change due to the continuous make-break action of
the starter. The inductor opposes this change by the
@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 1 | Nov-Dec 2017
Page: 53
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
production of high voltage impulse. The capacitor
plays its part in trying to oppose this rising voltage
and tries to force a current through the choke. So the
entire phenomenon of induced voltage and charging
current is confined to the local circuit containing the
parallel combination of choke and capacitor. No high
voltage pulse is available for ionization. Hence
ionization is not possible and the fluorescent lamp
fails to start.
Another way is to allow the tube light to start
normally (without any capacitor across the choke). As
the tube becomes lit, the capacitor can be put in to the
circuit. This requires an additional switch. But as soon
as the capacitor is connected, the light output of the
lamp decreases drastically. This occurs due to
increase in voltage drop across the choke. Now if a
load is put across the choke the voltage drop across
the choke decreases and the light output increases.
This load can be increased till the lamp starts
operating with its normal lumen output. With the
capacitor put across the choke, more loads can be put
across the choke than that without a capacitor. This is
a great method of extracting energy from the choke.
 Experiment no.6
After the tube light attained its normal operating state
a load was connected across the starter terminals. But
the observations made were quite contradictory from
the earlier experiments. While in the earlier
experiments connecting a load across the choke made
the tube brighter, connecting the same across the
starter made the tube dimmer. A 100W lamp
completely extinguished the tube light. Connecting a
load across the choke and the starter at the same time
revealed an interesting fact. When 100W load is put
across the choke and a 60W load is put across the
starter, the tube remains alight. But when a 60W lamp
is put across the choke and a 100W across the starter
the tube gets extinguished. When one lamp is
connected, the glow of the other decreases. But on
connecting a 2.5µF capacitor across the starter, the
glow of the lamp connected across the choke
increases.
 Experiment no.7
The variation of operating power factor with variation
in connected load across the choke was plotted in a
curve.
Graph no.5
As is seen from the curve the power factor increases
almost linearly with increase in load put across the
choke. The extrapolated trend line shows that unity
power factor will be reached at a load of around
160W. But such a huge load can not be normally
connected because it would damage the filament of
the tube light by increasing the voltage across it. So it
is self explicit that unity power factor can not be
reached. But magnetic ballast has an ‘unaided’
working power factor of .5 as supplied by the
manufacturers. This can be easily increased to .7 (or
close to .8 in some cases) by connecting a suitable
load across the choke.
@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 1 | Nov-Dec 2017
Page: 54
International Journal of Trend in Scientific Research and Development (IJTSRD) ISSN: 2456-6470
 Experiment no. 8
In this experiment a rheostat was connected in series
with the tube light supply in order to vary the supply
voltage to the device. A suitable load was connected
across the choke. The voltage was varied from zero to
full value (230V). For resistive loads such as
incandescent lamps, the light struck even at lower
voltage of 200V. The striking ability reduced with
increase in load. The similar effect was noticed in
case of an inductive load. But capacitor did not allow
the light to strike. Only a whitish glow at the two ends
of the tube was visible.
Observations: Energy can be extracted from the
choke of a tube light in two forms viz. D.C and A.C.
D.C nowadays is not used for domestic purposes.
Loads operated from the choke operate at a lower
value of voltage compared to their rated one and
hence don’t produce maximum output. However a fan
may be made to operate at a lower voltage instead of
using a regulator. A television set may be operated in
the same way because it works over a large range of
voltage. Almost 20% energy is saved is such cases.
But on using a rotating load such a fan the tube light
shows a tendency to flicker. Putting in another way,
the stroboscopic effect is much more pronounced.
However it is not advisable to put loads >100W
across the choke. Battery chargers work very
efficiently when connected across the choke. Under
any circumstance the following occur:—
a. Lumen output of the tube light would increase.
b. Power factor would be improved.
The loads esp. if they are inductive must be connected
only after the tube light has struck. Otherwise it may
not strike at all (explained earlier).
Results and Discussions:
The important results may be summarized as follows:
A. A load may be connected across an inductive
choke of a fluorescent lamp to extract a part of
the energy wasted in the choke.
B. The load should be preferably resistive, such as
incandescent lamps.
C. The connected load has a limited value beyond
which it can not be increased to avoid filament
damage of the tube light.
D. The power factor improves to an extent but does
not reach unity.
E. In general the load across the choke must be
connected after the fluorescent lamp assumes its
normal operation.
F. A capacitor connected across the choke does not
allow the lamp to strike.
G. A capacitor may be connected to the choke only
after the tube assumes its normal operation. In
this case the load that may be put across a choke
has a higher value than the one which may be
connected without using a capacitor.
H. Whenever a load is connected across the choke,
the lumen output of the lamp increases. However
this is not the case if only a capacitor is
connected across the choke. In this case the
lumen output drastically decreases.
I. Since the voltage across the choke load
combination is less than 230V (the voltage rating
of most single phase loads), the load does not
operate at its rated capacity.
J. The load may be connected across the starter
terminals after the tube light has lit up. But it has
a tendency of extinguishing the tube light.
K. The load that may be put across a starter is much
less than that which may be put across a choke.
So extraction of energy from the starter terminals
is only possible for loads ≤15W. This is not an
efficient method of energy extraction. As the
permissible load is small, the power factor does
not show much improvement.
So instead of replacing an old tube light with a new
one, we can easily connect a suitable load across the
choke. This would increase the light output as well as
the power factor. In this way discarding of magnetic
ballast may also be avoided.
Conclusion:
Fluorescent lamps which are common sources of
providing light in homes conventionally operate using
a choke and a starter. The low power factor and the
energy wastage problems may be taken care of by
connecting a suitable load across the choke. Loads
such as televisions operate over a wide range of
voltage generally show normal operation unless the
voltage drops below 100V. So a T.V may also be
connected across the choke. This is particularly useful
for old tube lights whose light tube light output has
deteriorated with aging. Instead of discarding the old
tube light, a load may be connected across the choke
to improve its performance. The method is quite
simple, inexpensive and easy to implement.
Acknowledgements:
This research has been financed and constantly
encouraged by Mr. P.B Nandi and Mrs. K. Nandi.
@ IJTSRD | Available Online @ www.ijtsrd.com | Volume – 2 | Issue – 1 | Nov-Dec 2017
Page: 55